Retention of visitors coming
to Palco 3.0 via the news in
SAPO:
Decoy optimization
 Readers of SAPO (www.sapo.pt) are presented with
Palco news.
 News appear occasionally that redirect the user to
PalcoPrincipal (www.palcoprincipal.pt).
 Problem: people who come to Palco in this way
usually don't stay there long enough.
 Question: How can we help Palco to retain them?
Introduction
http://sapo.pt
PalcoPrincipal article
http://palcoprincipal.pt
Latest news
+ Related news
+ Tags
http://palcoprincipal.pt
Proposal
 Task: improve current level of bouncing of users
(users coming to www.palcoprincipal.pt through news)
 Hypothesis: Shared interests are more common
among the users who clicked on a specific link (are
viewing the same content) than among users who
didn't (are not).
 The link clicked is usually the only information we
have about users coming from www.sapo.pt
Another approach
 Exploiting the social network
 Hypothesis: socially related users share interests
 e.g. we can identify users of Palco who showed
interest in particular items and use their activity to
make suggestions to socially related users
 Can be used for increasing fidelity of users who are
already registered on Palco
Data ­ statistics
Query No of results
users 68314
listeners 50663 (74%)
artists 17660 (26%)
friendship ties 168342 (avg 2.5)
mainstream bands 114062
fans 42228
comments 817370
tags 55018
Data ­ users
 User, User_profile
 LastLogin
 Genre ­ user's favourite genre (rock, r'n'b, ...)
 Name
 Nickname
 Gender
 BirthDate
 Type ­ type of the user profile (artist, listener, ...)
 Culture ­ user's nationality (e.g. Portuguese, Brazilian, ...)
 About ­ free text written by the user
 CreatedAt
 UpdatedAt
Data ­ contact, friends
 Contact
 Country
 City
 Place ­ free text
 Address
 PhoneNumber
 Homepage
 Friends
 InviterUserId ­ user who sent the friend request
 InvitedUserId ­ user who received the friend request
 CreatedAt
 UpdatedAt
Data ­ activities
 Owner_activities
 Action
 Owner
 Target
 Creator
 CreatedAt
 Slug
 Activity_stream_action
 ActionTag ­ e.g. new event added, band music added, ...
 PrivacyTagTitle
 PrivacySubscriberTagTitle
 ActionTitle
Data ­ tags, comments
 Tag
 Culture ­ e.g. Portuguese, Brazilian
 Name ­ name of the tag
 Slug ­ nickname for the tag
 Length ­ number of characters
 CreatedAt
 Comments
 Author ­ author of the comment
 OwnerId ­ owner of the comment
 CommentableModel ­ e.g. listener, track, photo, ...
 Text ­ text of the comment
 CreatedAt
Data - music
 Playlist
 Listener ­ owner of the playlist
 Name
 Description
 CreatedAt
 UpdatedAt
 Tracks ­ tracks contained in the playlist
 Track
 Album
 Name
 Downloadable
Data ­ mainstream bands
 MainstreamBand
 Summary
 About ­ somewhat longer than summary
 Name
 CreatedAt
 UpdatedAt
 LastFmLink ­ http://last.fm url
 Fans
 mainstreamBand
 user
Data ­ artists, albums
 Artist
 MusicUploads ­ number of uploaded tracks
 GooglePageRank
 GooglePageRankUpdatedAt
 Albums
 Artist
 Name
 Description
 ReleaseDate
 CreatedAt
 UpdatedAt
What can be done?
 Assortative mixing (homophily)
 Properties of the social network (six degrees of
separation, power law distribution)
 Cohesive group detection
 Visual analysis
 ???